Method and apparatus for classifying signals method and apparatus for generating descriptors and method and apparatus for retrieving signals
作者： Masayuki Nishiguchi , Mototsugu Abe
DOI:
关键词:
摘要: The input signal can be quickly and accurately classified a descriptor generated according to the result of classification. Then, retrieved on basis classification or descriptor. A processing apparatus comprises time block splitting section 3 for an audio into blocks that are typically 1 second long, feature extracting 4 characteristic quantity 18 degrees attribute from in each vector quantizing 5 carrying out operation categorical by means quantization technique uses VQ code book 8 formed degrees. outputs label obtained as indicating reliability label. If retrieving is conducted downstream, used retrieval.
google.de 参考文章(7)
Ravi Chandran, Bruce E. Dunne, Daniel J. Marchok, Speech presence measurement detection techniques ,(2001)
Ravi Chandran, Communication system noise cancellation power signal calculation techniques Journal of the Acoustical Society of America. ,vol. 121, pp. 687- ,(2001) , 10.1121/1.2640166
Shuwu Wu, John Mantegna, Audio codec using adaptive sparse vector quantization with subband vector classification Journal of the Acoustical Society of America. ,vol. 108, pp. 886- ,(1997) , 10.1121/1.1319427
Eric Lindemann, Musical synthesizer capable of expressive phrasing The Journal of the Acoustical Society of America. ,vol. 117, pp. 2700- 2700 ,(2005) , 10.1121/1.1932401
Taro Suito, Toshiya Akiba, Naohisa Arai, Masashi Ohta, Noboru Murabayashi, Takao Takahashi, Masami Tomita, Apparatus and method for recording audio/video signal and detected feature information thereof ,(1997)
Nobumasa Seiyama, Tohru Takagi, Atsushi Imai, Adaptive speech rate conversion without extension of input data duration, using speech interval detection ,(1998)